Pedro Gómez-Meneses

Pedro Gómez-Meneses

Auto-generated profile, based on publicly available information

Pedro Gómez-Meneses

Bond University

Research Focus & Metrics

Academic Career Length

19Years (2008-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 July 2010
Verified
Conference Paper
Open access
PGPedro Gomez-Meneses
Pedro Gomez-Meneses
Marcus RandallMarcus Randall
Marcus Randall
ALAndrew Lewis
Andrew Lewis
Extremal optimisation (EO) is a relatively recent nature-inspired heuristic whose search method is especially suitable to solve combinatorial optimisation problems. To date, most of the research in EO has been applied for solving single-objective problems and only a relatively small number of attempts to extend EO toward multi-objective problems. This paper presents a hybrid multi-objective version of EO (HMEO) to solve multi-objective combinatorial problems. This new approach consists of a multi-objective EO framework, for the coarse-grain search, which contains a novel multi-objective combinatorial local search framework for the fine-grain search. The chosen problems to test the proposed method are the multi-objective knapsack problem and the multi-objective quadratic assignment problem. The results show that the new algorithm is able to obtain competitive results to SPEA2 and NSGA-II. The non-dominated points found are well-distributed and similar or very close to the Pareto-front found by previous works.

1 July 2017 in IFAC-PapersOnLine
Verified
Journal Article
Closed access
Hugo Omar GarcésHugo Omar Garcés
Hugo Omar Garcés
JAJosé Abreu
José Abreu
PGPedro Gómez
Pedro Gómez
CCClaudia Carrasco
Claudia Carrasco
LALuis Arias
Luis Arias
AFAndrés Fuentes
Andrés Fuentes

1 January 2009 in Lecture Notes in Computer Science
Verified
Book Chapter
Closed access
PDF file
PGPedro Gómez-Meneses
Pedro Gómez-Meneses
Marcus RandallMarcus Randall
Marcus Randall



6 February 2026 in IEEE Access
Verified
Journal Article
Open access
ASAlejandra Segura-Navarrete
Alejandra Segura-Navarrete
CMClaudia Martinez-Araneda
Claudia Martinez-Araneda
SLSoledad Martinez Labrin
Soledad Martinez Labrin
PGPedro Gómez-Meneses
Pedro Gómez-Meneses
CVChristian Vidal-Castro
Christian Vidal-Castro
TGTatiana Gutiérrez-Bunster
Tatiana Gutiérrez-Bunster
Background: Respect and inclusion are essential for gender equity in higher education, where subtle forms of gender-based violence, such as bias and stereotypes, often emerge and affect academic well-being. The systematic use of machine learning or deep learning models to automatically identify gender bias in higher education teaching evaluations remains underexplored, providing the main motivation for the present study. Purpose: This work evaluates the ability of machine learning (ML) and deep learning (DL) models to identify textual expressions of gender bias on digital platforms in higher education. Using student evaluations of teaching from the University of BÍo-BÍo (Chile), we characterize bias patterns linked to instructor gender. We address two questions: RQ1: To what extent do student comments in university teaching evaluations written in Spanish exhibit explicit gender biases and stereotypical patterns when stratified by the instructor’s gender? and RQ2: How accurately do ML and DL models (SVM, BETO, DistilBERT) detect explicit gender bias in this domain-specific corpus when evaluated using macro-average metrics? Method: We conducted a PRISMA-guided systematic review (2018–2024) to identify suitable detection techniques, followed by a determinant-variable analysis examining how gender-related attributes, stereotypes, and comment features influence bias in feedback. Student evaluations from 2017–2019 and 2023 were annotated with binary labels and used to train and evaluate models. The dataset contains 960 comments (691 for training, 77 for validation, and 192 for testing). We then developed predictive models, including SVMs and transformer-based (BETO and DistilBERT). Results: For RQ1, an analysis of determinant variables shows that comments on female instructors’ teaching performance tend to emphasize personal traits and relational behaviors, particularly dedication, accessibility, and willingness to answer questions. In contrast, comments about male instructors more often highlight academic competence, with frequent mentions of subject mastery and content expertise, as well as expressions of admiration for their intellectual performance. For RQ2, BETO+BiLSTM and DistiIBERT+BiLSTM were developed, which performed strongly, achieving average macro Fl score of 99% on the gender bias detection task, well above the BETO and SVM baselines of 95% and 88%, respectively. As part of the results, a dataset of 960 comments labeled as gender-based bias (GB) and non-gender-based bias (N-GB) was compiled and made available on the Zenodo repository at https://doi.org/10.5281/zenodo.17665445